Predictive Analytics, Visualization Enlisted in Crime Fight

A visualization tool and predictive analytics package targeting law enforcement and public safety applications is said to be the first tool of its kind to combine real-time social media and Internet data feeds to help predict and reduce crime.

Hitachi Data Systems said Monday (Sept. 28) its visualization tool along with its “predictive crime analytics” supports the company’s smart city efforts. That initiative seeks to advance municipal public safety efforts through the application of predictive and other advanced analytics along with better access to video data. The Hitachi package also includes a video management platform.

Hitachi’s visualization suite is a hybrid cloud-based platform that integrates disparate data and video assets from public safety systems such as 911 computer-aided dispatch, license plate readers and gun shot sensors. These data are integrated in real time and then presented geospatially. The video platform provides police with analyses with the goal of improving intelligence, enhancing investigative capabilities and increasing operational efficiencies, the company claimed.

Along with capturing real-time event data from sensors, the video suite also provides geospatial visualizations for historical crime data in several forms, including heat maps. That feature is available in a predictive crime analytics add-on module to the 4.5 software release of the visualization tool.

The predictive analytics capability uses spatial and temporal algorithms to combine real-time event data from public safety cameras and other sensors with historical and contextual crime data stored in databases. It also integrates social media and other sources. The algorithms can be used to assign threat levels for every city block, Hitachi claimed. The algorithms can be used to assign threat levels so law enforcement or first responders can forecast where crimes are likely to occur and deploy resources accordingly.

Hitachi also promoted the visualization and analytics tool as the first predictive policing technique to use natural language processing for topic intensity modeling. That technique uses social media networks together with other public and private data feeds in real time to deliver crime predictions.

The accompanying appliance-based video management platform uses VMware software for data integration and visualization Hitachi-supported hardware separates storage and computing. The platform also supports most video management software to handle the storage and processing of different types of video data.

Police departments are increasingly turning to predictive analytics as they seek ways to deploy scarce public safety resources. Along with Hitachi’s visualization approach, other emerging companies like the big data analytics firm H₂O are trying to leverage a machine learning technology known as “deep learning” as a way to improve community policing. The approach also seeks to predict which crimes are most like to result in arrests and crime prevention.

Hitachi Data Systems said its visualization suite 4.5 with predictive crime analytics and its video management platform are generally available now.

Recent items:

What Police Can Learn From Deep Learning

Police Push the Limits of Big Data Technology